This invention relates to the inspection of surface topology, and in particular to inspection of the surfaces of structural members subject to high stresses.
In structural members subject to high stresses, for example metallic or fibre reinforced plastics composite wings, the topology of the surface of the member can be of vital importance. In metallic bolted structures, for example, cold expansion of fastener holes can increase the fatigue life of the member many times by prevention of the formation of cracks emanating from hole walls. A standard method of cold expanding such holes involves drawing a mandrel through a split sleeve within the hole and expanding the sleeve to form a zone of compression in the material surrounding the hole. The effect of creating this zone of compression around the hole is to cause a volcano-like disturbance or eruption of the surface surrounding the hole, due to the Poisson effect. This surface disturbance or eruption can extend up to a radius from the hole edge. Contrary to general belief it has now been found that the extent of the Poisson volcano around the cold-expanded hole is directly related to the material properties of the member and to the degree of expansion applied to the hole, at least when such expansion is carried out in a controlled manner for example by the above mentioned method of drawing a mandrel through a split sleeve. The degree of expansion, eg. 1%, 2%, 3%, 4% etc. determines the level of fatigue enhancement applied to the hole. A very fine balance exists however between causing damage to the member, eg. plate material, by over-expanding on the one hand, and providing insufficient fatigue protection by under expanding on the other. For this reason it is important to be able to ascertain not only the presence or absence of cold expansion of holes but also the degree of such cold expansion in order to guarantee the integrity of structural members such as those in modern aircraft structures.
In our U.S. Pat. No. 5,619,327 we disclose a method of detecting cold expansion of holes in such a structural member as an aircraft wing. Unfortunately this method whilst fulfilling a vital and pressing need in enabling the first of the two above requirements to be determined, does not disclose a method of achieving the second in a reliable and repeatable way.
In addition, in the context of highly stressed composite materials, the degree of damage caused to a structure by an object impacting a surface of the structure can be easily hidden by the phenomenon of barely visible impact damage. According to this phenomenon structural damage can actually increase with distance from a surface to which an impact has occurred. The ability to ascertain the extent of such damage by the inspection of the topology of the impacted surface would be extremely valuable because a technique is available using a comparison of surface damage area and ultrasonic C scan measurements to determine this.
It is an object of the invention to provide a method of inspecting the topology of a surface of a structural member to ascertain the degree of cold working of metal immediately surrounding holes penetrating that surface and also to ascertain the extent of damage to impacted composite structures.
According to the invention there is provided a method of inspecting the topology of a surface of a structural member to determine the degree to which a known type of stress has been applied to the member, the method including the steps of providing a range of calibration samples of structurally equivalent members, the samples each having been subject to the known type of stress to a differing respective degree; supporting a Moirxc3xa9 grid in a position spaced from and co-extending with the surface to be inspected and at a small included angle to the surface; directing a source of light through the grid to strike the surface at an oblique angle thereto; viewing the surface through the grid in a direction substantially normal to the surface to view interference fringes and regions of distortion to the fringes, and measuring the extent of a said region of distortion and comparing the said measurement with equivalent measurements taken for respective calibration samples to determine the degree of stress applied to the member.
By the term, xe2x80x9cstructurally equivalentxe2x80x9d is meant a sample having the same type of structure, eg. reinforced in the same way, the same structurally relevant dimensions, eg. thickness of the sample, and being of the same material. The interference fringes comprise shadows of the grid projected onto the surface combined with lines of the grid as directly viewed by the observer, and the regions of distortion of the fringes correspond to disturbance to the topology of the surface caused by a said application of stress to the member acting via the Poissons effect.
The step of making a said comparison between the said measurement and equivalent distances measured for calibration samples of a member of the same type, relevant dimensions (for example thickness) and material, may include determining which sample exhibits substantially the same measured distance as the said member, and noting the degree of stress applied to that sample, which degree of stress will substantially correspond to the stress, eg. degree of cold working of a hole, applied to the said member.
The method may include the step of adjusting the small included angle, for example by using a micrometer adjuster, to provide substantially the same fringe density as for the equivalent measurements taken on the calibration samples. In this way consistency of measurement is ensured.
The line spacings of the Moirxc3xa9 grid are preferably substantially in the range 1 to 200 lines per mm. At a line spacing of 1 per mm the apparatus would be capable of detecting a major dent, for example, in a surface of a member being inspected. Apparatus having line spacing within the range of substantially 5 to 20 lines per mm is preferred and a line spacing of substantially 10 lines per mm is most preferred, giving good resolution and range of application.
The method preferably includes the step of providing a said source of light of a generally parallel nature, eg. from a collimated source, to provide a linear interference pattern so as to facilitate accurate measurements when the observer""s eye does not quite view the surface normal thereto.
The step of providing the source of light may conveniently comprise providing a substantially point source of light, eg. a light emitting diode. Provided the observer views the surface substantially normal thereto little loss of accuracy in measurement will result when such a light source is used.
The step of viewing the surface substantially normal thereto may be ensured by providing an optical aid so to indicate to the observer. The optical aid may comprise a mirror facing the observer fast with the body, the mirror having a line indicator spaced therefrom in the direction of the observer whereby, when the observer views the surface substantially normal thereto, no reflection of the line indicator will be observable in the mirror.
The step of measuring the extent of a said region of distortion may comprise taking the greatest possible measurement across the region substantially in a direction of undistorted fringes from commencement of distortion on one side thereof to ending of distortion on the other side thereof.
Where the distortion to the surface comprises an annular region surrounding a hole in the surface, the step of measuring the extent of the region of distortion may comprise taking the measurement substantially in a direction of undistorted fringes from a periphery of a said hole to a point of fringe distortion farthest from the said periphery.
The said oblique angle is desirably either 45 degrees or 63.2 degrees to enable a convenient relationship between grating pitch and surface displacement to be formed. The angles of 45 degrees and 63.2 degrees provide particularly convenient relationships between fringe order, grating pitch and displacement resolution since the vertical displacement per fringe is given by the reciprocal of the grating pitch divided by the tangent of the included angle. Thus tan 45=1, giving a resolution of 1 per pitch and tan 63.2 is approximately equal to 2, giving twice the resolution.
The method may include the step of providing a sighting device, for example a graticule, for the observer, which sighting device may be centred upon a region of distortion, eg. the centre of a hole believed to be cold worked, to facilitate accurate taking of the said measurements.
The observer may view the surface via the medium of a camera so that images of the measurements taken may be recorded. And more preferably the camera comprises a high resolution, preferably digital, still camera to enable the best possible accuracy of measurement to be achieved, when recording. When a digital camera is used, results of measurements taken may conveniently be transferred to a computer, for example for image processing or for long term use. Such a computer may be programmed to search for particular shapes of curve representing commencement or ending of regions of distortion of the fringes. In this way consistency of measurement between calibration samples and the member under test may be better achieved.